Notebook metadata changed
xxxxxxxxxx
29
 
1
{
2
  "accelerator": "GPU",
3
  "colab": {
4
    "collapsed_sections": [
5
    ],
6
    "provenance": [
7
    ]
8
  },
9
  "interpreter": {
10
    "hash": "e64f7e1a97f549196798c7884a74f44ae7cc56a613fb63e093872097d3605f11"
11
  },
12
  "kernelspec": {
13
    "display_name": "Python 3 (ipykernel)",
14
​
⇛⇚
xxxxxxxxxx
19
 
1
{
2
  "kernelspec": {
3
    "display_name": "Python 3 (ipykernel)",
4
​

2022-09-15¶

Metadata changed
xxxxxxxxxx
3
 
1
{
2
  "id": "W_5ZhvIqpdat"
3
}
⇛⇚
xxxxxxxxxx
3
 
1
{
2
​
3
}
xxxxxxxxxx
1
 
1
[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/SzabolcsWeyde/DeepLearningHW2/blob/b48faedc80c15318e090e7878b297daa7eb59bbf/DeepLearningHW2.ipynb)
Metadata changed
xxxxxxxxxx
In [1]:
In [3]:
xxxxxxxxxx
1
 
1
%matplotlib inline
Metadata changed
xxxxxxxxxx
3
 
1
{
2
  "id": "HOU2IE2upfX6"
3
}
⇛⇚
xxxxxxxxxx
3
 
1
{
2
​
3
}
In [2]:
In [4]:
xxxxxxxxxx
1
 
1
import matplotlib.pyplot as plt
Metadata changed
xxxxxxxxxx
3
 
1
{
2
  "id": "wGxlOrVkqbkJ"
3
}
⇛⇚
xxxxxxxxxx
3
 
1
{
2
​
3
}
In [3]:
In [5]:
xxxxxxxxxx
1
 
1
import numpy as np
Metadata changed
xxxxxxxxxx
3
 
1
{
2
  "id": "0_iemkZbqkwm"
3
}
⇛⇚
xxxxxxxxxx
3
 
1
{
2
​
3
}
In [4]:
In [6]:
xxxxxxxxxx
1
 
1
! pip install tensorboardX
Metadata changed
xxxxxxxxxx
7
 
1
{
2
  "colab": {
3
    "base_uri": "https://localhost:8080/"
4
  },
5
  "id": "VbJ20rZBqrWe",
6
  "outputId": "6db51a37-c246-4423-c197-6160dd495348"
7
}
⇛⇚
xxxxxxxxxx
3
 
1
{
2
​
3
}
Outputs unchanged
Requirement already satisfied: tensorboardX in c:\users\weysz\anaconda3\envs\deeplearning\lib\site-packages (2.5.1)
Requirement already satisfied: numpy in c:\users\weysz\anaconda3\envs\deeplearning\lib\site-packages (from tensorboardX) (1.22.3)
Requirement already satisfied: protobuf<=3.20.1,>=3.8.0 in c:\users\weysz\anaconda3\envs\deeplearning\lib\site-packages (from tensorboardX) (3.20.1)
application/vnd.jupyter.stdout
In [5]:
In [7]:
xxxxxxxxxx
1
 
1
%load_ext tensorboard
Metadata changed
xxxxxxxxxx
3
 
1
{
2
  "id": "VuqByoGbqu3s"
3
}
⇛⇚
xxxxxxxxxx
3
 
1
{
2
​
3
}
In [6]:
In [8]:
xxxxxxxxxx
1
 
1
from tensorboardX import SummaryWriter
Metadata changed
xxxxxxxxxx
3
 
1
{
2
  "id": "f9ivkkycq255"
3
}
⇛⇚
xxxxxxxxxx
3
 
1
{
2
​
3
}
In [7]:
In [9]:
xxxxxxxxxx
1
 
1
from datetime import datetime
Metadata changed
xxxxxxxxxx
3
 
1
{
2
  "id": "g2M1YZpNq8FE"
3
}
⇛⇚
xxxxxxxxxx
3
 
1
{
2
​
3
}
In [8]:
In [10]:
xxxxxxxxxx
1
 
1
logdir = "logs"
Metadata changed
xxxxxxxxxx
3
 
1
{
2
  "id": "E8NiIWYWq_pD"
3
}
⇛⇚
xxxxxxxxxx
3
 
1
{
2
​
3
}
In [9]:
In [11]:
xxxxxxxxxx
2
 
1
def activation(x):
2
  return 1.0 / (1.0 + np.exp(-x))
Metadata changed
xxxxxxxxxx
3
 
1
{
2
  "id": "4IjWzavRrBmb"
3
}
⇛⇚
xxxxxxxxxx
3
 
1
{
2
​
3
}
In [10]:
In [12]:
xxxxxxxxxx
2
 
1
def dactivation(x):
2
  return np.exp(-x)/(1.0 + np.exp(-x))**2
Metadata changed
xxxxxxxxxx
3
 
1
{
2
  "id": "7OVFaefprnZq"
3
}
⇛⇚
xxxxxxxxxx
3
 
1
{
2
​
3
}
In [11]:
In [13]:
xxxxxxxxxx
167
 
1
​
36
      self.layers[i][...] =  activation(s_i)
37
    return self.layers[-1]
38
​
39
#############################################################################
40
# I have separated the backward method. The calc_deltas method calculates   #
41
# the gradients and the update_weigths method updates the weigths           #
42
#############################################################################
43
​
44
  def calc_deltas(self,target,lrate=0.1):
45
    deltas = []
46
    dws = []
47
    derror = -(target-self.layers[-1])
48
    
49
    s_last = np.dot(self.layers[-2],self.weights[-1])
50
    delta_last = derror *  dactivation(s_last)
51
    deltas.append(delta_last)
52
    
53
    for i in range(len(self.shape)-2,0,-1):
54
      s_i = np.dot(self.layers[i-1],self.weights[i-1])
55
      delta_i = np.dot(deltas[0],self.weights[i].T)*dactivation(s_i)
56
      deltas.insert(0,delta_i)
57
​
58
    for i in range(len(self.weights)):
59
      layer = np.atleast_2d(self.layers[i])
60
      delta = np.atleast_2d(deltas[i])
61
      dw = -lrate*np.dot(layer.T,delta)
62
      dws.append(dw)
63
    return dws
64
​
65
  def update_weigths(self,target,dws,lrate=0.1):
66
​
67
    for i in range(len(self.weights)):
68
      self.weights[i]+=dws[i]
69
      self.dw[i]= dws[i]
70
    error = (target -  self.layers[-1])**2
71
    return error.sum()
72
    
73
##############################################################################
74
​
75
  def propagate_backward(self,target,lrate=0.1):
76
    deltas = []
77
​
90
      layer = np.atleast_2d(self.layers[i])
91
      delta = np.atleast_2d(deltas[i])
92
      dw = -lrate*np.dot(layer.T,delta)
93
      self.weights[i]+=dw
94
​
95
​
98
​
99
from sklearn import preprocessing
100
def learn(network,X,Y,valid_split,test_split,write,epochs=20,lrate=0.1,batch_size=1):
101
​
102
  nb_samples = len (Y)
103
​
115
  randperm = np.random.permutation(len(X_train))
116
  X_train,Y_train = X_train[randperm],Y_train[randperm]
117
########################################################################
118
# I have added a for loop, that calculates the gradient of the sampels #
119
# in a mini batch. After the loop, I calculate the average of the      #
120
# gradintes, and then update the wieghts                               #
121
########################################################################
122
​
123
  for i in range (epochs):
124
    train_err = 0
125
    dws=[]
126
    deltas=[]
127
    for k in range (int(X_train.shape[0]/batch_size)):
128
      for j in range(batch_size):
129
        network.propagate_forward(X_train[k* batch_size +j])
130
        deltas=network.calc_deltas(Y_train[k* batch_size +j])
131
        dws.append(deltas)
132
​
133
      deltas=[]
134
      for l in range(len(dws[0])):
135
        d=np.mean([item[l] for item in dws],axis=0)
136
        deltas.append(d)
137
      train_err +=network.update_weigths(Y_train[k* batch_size +j],deltas)*batch_size
138
      dws=[]
139
​
140
#########################################################################
141
    train_err /=X_train.shape[0]
142
​
143
​
159
    test_err += (o_test[k]-Y_test[k])**2
160
  test_err /= X_test.shape[0]
161
  print(test_err)            
162
    
163
  
164
  
165
​
166
​
167
​
⇛⇚
xxxxxxxxxx
108
 
1
​
36
      self.layers[i][...] =  activation(s_i)
37
    return self.layers[-1]
38
  def propagate_backward(self,target,lrate=0.1):
39
    deltas = []
40
​
53
      layer = np.atleast_2d(self.layers[i])
54
      delta = np.atleast_2d(deltas[i])
55
​
56
      dw = -lrate*np.dot(layer.T,delta)
57
​
58
      self.weights[i]+=dw
59
​
60
​
63
​
64
from sklearn import preprocessing
65
def learn(network,X,Y,valid_split,test_split,write,epochs=20,lrate=0.1):
66
​
67
  nb_samples = len (Y)
68
​
80
  randperm = np.random.permutation(len(X_train))
81
  X_train,Y_train = X_train[randperm],Y_train[randperm]
82
​
83
  for i in range (epochs):
84
    train_err = 0
85
    for k in range (X_train.shape[0]):
86
      network.propagate_forward(X_train[k])
87
      train_err +=network.propagate_backward(Y_train[k],lrate)
88
    train_err /=X_train.shape[0]
89
​
90
​
106
    test_err += (o_test[k]-Y_test[k])**2
107
  test_err /= X_test.shape[0]
108
  print(test_err)              
Metadata changed
xxxxxxxxxx
3
 
1
{
2
  "id": "VscARWeJr8xH"
3
}
⇛⇚
xxxxxxxxxx
3
 
1
{
2
​
3
}
In [12]:
In [14]:
xxxxxxxxxx
1
 
1
network = MLP (2,10,1)
Metadata changed
xxxxxxxxxx
3
 
1
{
2
  "id": "3vjxJtVl-VxI"
3
}
⇛⇚
xxxxxxxxxx
3
 
1
{
2
​
3
}
In [13]:
In [15]:
xxxxxxxxxx
9
 
1
nb_samples=1_000
2
X = np.zeros((nb_samples,2))
3
Y = np.zeros(nb_samples)
4
​
⇛⇚
xxxxxxxxxx
9
 
1
nb_samples=1000
2
X = np.zeros((nb_samples,2))
3
Y = np.zeros(nb_samples)
4
​
Metadata changed
xxxxxxxxxx
3
 
1
{
2
  "id": "W_yUMRf1-b1V"
3
}
⇛⇚
xxxxxxxxxx
3
 
1
{
2
​
3
}
In [16]:
In [16]:
xxxxxxxxxx
13
 
1
​
8
network.reset()
9
valid_split = 0.2; test_split = 0.1
10
time=now = datetime.now()
11
learn(network, X, Y, valid_split, test_split, writer, 500,lrate=0.1,batch_size=16)
12
time2= now = datetime.now()
13
print(f'elapsed time:{time2-time}')
⇛⇚
xxxxxxxxxx
10
 
1
​
8
network.reset()
9
valid_split = 0.2; test_split = 0.1
10
learn(network, X, Y, valid_split, test_split, writer, 100)
Metadata changed
xxxxxxxxxx
7
 
1
{
2
  "colab": {
3
    "base_uri": "https://localhost:8080/"
4
  },
5
  "id": "Y-NWIm0J_Jy7",
6
  "outputId": "c4493c42-9de6-4d50-a703-c6057cb52f3a"
7
}
⇛⇚
xxxxxxxxxx
3
 
1
{
2
​
3
}
Outputs changed
Output added
0 epoch, train_err: 0.24614314439761184, valid_err: 0.23784604654950872
1 epoch, train_err: 0.2297025824522868, valid_err: 0.2211767844090273
2 epoch, train_err: 0.20695727395811084, valid_err: 0.19340898826237177
3 epoch, train_err: 0.17415096809613942, valid_err: 0.15847635538727112
4 epoch, train_err: 0.13874775481191173, valid_err: 0.12667624882747902
5 epoch, train_err: 0.10981930239183453, valid_err: 0.10346711821683872
6 epoch, train_err: 0.08970630277330417, valid_err: 0.08786103487832059
7 epoch, train_err: 0.0763555883214336, valid_err: 0.07733361232650038
8 epoch, train_err: 0.06734767241493564, valid_err: 0.06998772032137736
9 epoch, train_err: 0.061033865171153, valid_err: 0.06465801307469875
10 epoch, train_err: 0.05642294582752031, valid_err: 0.06064978918367313
11 epoch, train_err: 0.05292709294356913, valid_err: 0.057540681448743865
12 epoch, train_err: 0.05018970754043007, valid_err: 0.05506522152980132
13 epoch, train_err: 0.04798708867732972, valid_err: 0.05305060002673568
14 epoch, train_err: 0.046173904089914626, valid_err: 0.05138053581203745
15 epoch, train_err: 0.044652548066732516, valid_err: 0.0499743932291983
16 epoch, train_err: 0.043355447156445334, valid_err: 0.04877472564827305
17 epoch, train_err: 0.04223451998777812, valid_err: 0.04773960860709058
18 epoch, train_err: 0.04125470130409357, valid_err: 0.046837779329531995
19 epoch, train_err: 0.04038984498800343, valid_err: 0.04604547031277881
20 epoch, train_err: 0.03962006121683461, valid_err: 0.04534429472026072
21 epoch, train_err: 0.0389299425744552, valid_err: 0.04471980209812429
22 epoch, train_err: 0.038307355630045135, valid_err: 0.04416047168686615
23 epoch, train_err: 0.03774260086896862, valid_err: 0.04365699775255655
24 epoch, train_err: 0.037227817843271296, valid_err: 0.043201773732828096
25 epoch, train_err: 0.03675655682891004, valid_err: 0.04278851422055424
26 epoch, train_err: 0.03632346558938465, valid_err: 0.0424119740887381
27 epoch, train_err: 0.03592405701840421, valid_err: 0.04206773709384292
28 epoch, train_err: 0.035554534459310484, valid_err: 0.04175205483857911
29 epoch, train_err: 0.035211658715623, valid_err: 0.041461722680489234
30 epoch, train_err: 0.03489264557620464, valid_err: 0.04119398304858718
31 epoch, train_err: 0.03459508593641429, valid_err: 0.040946449305073226
32 epoch, train_err: 0.034316882836875465, valid_err: 0.0407170451612329
33 epoch, train_err: 0.03405620130297326, valid_err: 0.04050395598333687
34 epoch, train_err: 0.03381142796987976, valid_err: 0.04030558927482347
35 epoch, train_err: 0.03358113826372626, valid_err: 0.04012054230839786
36 epoch, train_err: 0.033364069475584596, valid_err: 0.0399475753828193
37 epoch, train_err: 0.03315909847630221, valid_err: 0.039785589547158114
38 epoch, train_err: 0.032965223121668795, valid_err: 0.039633607907338206
39 epoch, train_err: 0.03278154661995979, valid_err: 0.03949075983211762
40 epoch, train_err: 0.032607264299455756, valid_err: 0.03935626752709621
41 epoch, train_err: 0.03244165233755781, valid_err: 0.03922943455941304
42 epoch, train_err: 0.03228405810671799, valid_err: 0.03910963600230503
43 epoch, train_err: 0.032133891863551346, valid_err: 0.038996309934796504
44 epoch, train_err: 0.03199061956200529, valid_err: 0.038888950082696126
45 epoch, train_err: 0.031853756613549396, valid_err: 0.03878709942661521
46 epoch, train_err: 0.03172286245011822, valid_err: 0.038690344633721285
47 epoch, train_err: 0.03159753577127084, valid_err: 0.03859831119446169
48 epoch, train_err: 0.03147741037741577, valid_err: 0.03851065916509197
49 epoch, train_err: 0.03136215150723691, valid_err: 0.038427079432644534
50 epoch, train_err: 0.03125145261058731, valid_err: 0.038347290431844
51 epoch, train_err: 0.03114503249879487, valid_err: 0.0382710352540416
52 epoch, train_err: 0.03104263282307182, valid_err: 0.038198079096995165
53 epoch, train_err: 0.03094401583895335, valid_err: 0.03812820701162135
54 epoch, train_err: 0.0308489624207027, valid_err: 0.038061221907981904
55 epoch, train_err: 0.03075727029465813, valid_err: 0.03799694278794698
56 epoch, train_err: 0.030668752464744726, valid_err: 0.037935203176378385
57 epoch, train_err: 0.030583235806965797, valid_err: 0.03787584972643234
58 epoch, train_err: 0.030500559812752865, valid_err: 0.03781874097779145
59 epoch, train_err: 0.03042057546366478, valid_err: 0.037763746249399224
60 epoch, train_err: 0.030343144222174472, valid_err: 0.0377107446506467
61 epoch, train_err: 0.030268137125208417, valid_err: 0.037659624197010876
62 epoch, train_err: 0.0301954339687731, valid_err: 0.03761028101792257
63 epoch, train_err: 0.0301249225734427, valid_err: 0.037562618646174745
64 epoch, train_err: 0.03005649812173212, valid_err: 0.037516547379518095
65 epoch, train_err: 0.02999006255946536, valid_err: 0.037471983706246624
66 epoch, train_err: 0.029925524054193856, valid_err: 0.03742884978758423
67 epoch, train_err: 0.0298627965045421, valid_err: 0.03738707299055854
68 epoch, train_err: 0.029801799095077998, valid_err: 0.037346585465811655
69 epoch, train_err: 0.02974245589193185, valid_err: 0.03730732376546254
70 epoch, train_err: 0.0296846954749381, valid_err: 0.03726922849671872
71 epoch, train_err: 0.029628450602558785, valid_err: 0.037232244007438686
72 epoch, train_err: 0.02957365790626305, valid_err: 0.037196318100293835
73 epoch, train_err: 0.029520257611417872, valid_err: 0.03716140177256429
74 epoch, train_err: 0.029468193282063296, valid_err: 0.037127448978943864
75 epoch, train_err: 0.02941741158723794, valid_err: 0.037094416415028135
76 epoch, train_err: 0.029367862086772202, valid_err: 0.03706226331941988
77 epoch, train_err: 0.02931949703469012, valid_err: 0.03703095129261706
78 epoch, train_err: 0.029272271198558904, valid_err: 0.03700044413104987
79 epoch, train_err: 0.029226141693299026, valid_err: 0.036970707674811615
80 epoch, train_err: 0.02918106782812431, valid_err: 0.03694170966778565
81 epoch, train_err: 0.029137010965415044, valid_err: 0.036913419629007
82 epoch, train_err: 0.02909393439045362, valid_err: 0.036885808734221426
83 epoch, train_err: 0.02905180319105591, valid_err: 0.03685884970671238
84 epoch, train_err: 0.02901058414623054, valid_err: 0.03683251671656089
85 epoch, train_err: 0.02897024562308269, valid_err: 0.03680678528758884
86 epoch, train_err: 0.028930757481256963, valid_err: 0.03678163221131315
87 epoch, train_err: 0.02889209098427797, valid_err: 0.03675703546730114
88 epoch, train_err: 0.028854218717215278, valid_err: 0.036732974149382
89 epoch, train_err: 0.02881711451014568, valid_err: 0.03670942839721786
90 epoch, train_err: 0.02878075336694183, valid_err: 0.036686379332788316
91 epoch, train_err: 0.0287451113989552, valid_err: 0.036663809001383225
92 epoch, train_err: 0.028710165763202476, valid_err: 0.03664170031673644
93 epoch, train_err: 0.028675894604700436, valid_err: 0.03662003700996755
94 epoch, train_err: 0.02864227700262523, valid_err: 0.036598803582028584
95 epoch, train_err: 0.02860929292000045, valid_err: 0.0365779852593806
96 epoch, train_err: 0.028576923156645024, valid_err: 0.036557567952648085
97 epoch, train_err: 0.02854514930513486, valid_err: 0.03653753821802385
98 epoch, train_err: 0.028513953709553327, valid_err: 0.036517883221214215
99 epoch, train_err: 0.028483319426824433, valid_err: 0.03649859070373329
TESZT
0.018471787966945822
application/vnd.jupyter.stdout
Output deleted
0 epoch, train_err: 0.28679125664959304, valid_err: 0.26342247385940903
1 epoch, train_err: 0.2604374085132547, valid_err: 0.2537258459410452
2 epoch, train_err: 0.25094027856141005, valid_err: 0.2514559195934984
3 epoch, train_err: 0.24761124563377193, valid_err: 0.25093219586758503
4 epoch, train_err: 0.24620476935152347, valid_err: 0.2507343324091377
5 epoch, train_err: 0.24544652693106767, valid_err: 0.25057919329925765
6 epoch, train_err: 0.24494036213144837, valid_err: 0.25042184042174076
7 epoch, train_err: 0.2445467110139412, valid_err: 0.25025970376085505
8 epoch, train_err: 0.24420995003693147, valid_err: 0.25009515448714476
9 epoch, train_err: 0.24390569896910413, valid_err: 0.24992978366274918
10 epoch, train_err: 0.2436224180881737, valid_err: 0.2497642076477498
11 epoch, train_err: 0.24335418853123617, valid_err: 0.24959848552817443
12 epoch, train_err: 0.24309765284255097, valid_err: 0.24943241250142242
13 epoch, train_err: 0.24285066004920824, valid_err: 0.24926567240080794
14 epoch, train_err: 0.24261165046117752, valid_err: 0.2490979087764192
15 epoch, train_err: 0.242379371398424, valid_err: 0.24892875560942315
16 epoch, train_err: 0.24215274329297037, valid_err: 0.24875784923776206
17 epoch, train_err: 0.24193079474904652, valid_err: 0.2485848318686294
18 epoch, train_err: 0.241712629477108, valid_err: 0.24840935147733273
19 epoch, train_err: 0.24149740812212064, valid_err: 0.24823106026234062
20 epoch, train_err: 0.24128433718284964, valid_err: 0.2480496126158072
21 epoch, train_err: 0.24107266141997313, valid_err: 0.2478646630193918
22 epoch, train_err: 0.24086165807757698, valid_err: 0.24767586402891886
23 epoch, train_err: 0.2406506321279249, valid_err: 0.2474828644029452
24 epoch, train_err: 0.2404389121575269, valid_err: 0.24728530738412938
25 epoch, train_err: 0.2402258467018617, valid_err: 0.2470828291239817
26 epoch, train_err: 0.24001080092503815, valid_err: 0.24687505723560296
27 epoch, train_err: 0.23979315358343678, valid_err: 0.24666160945833454
28 epoch, train_err: 0.23957229423380352, valid_err: 0.24644209241972515
29 epoch, train_err: 0.23934762065777196, valid_err: 0.24621610048247267
30 epoch, train_err: 0.23911853648158837, valid_err: 0.24598321466643427
31 epoch, train_err: 0.2388844489743343, valid_err: 0.2457430016381689
32 epoch, train_err: 0.23864476701128406, valid_err: 0.24549501276269622
33 epoch, train_err: 0.23839889919173682, valid_err: 0.2452387832142178
34 epoch, train_err: 0.23814625210298235, valid_err: 0.24497383114443397
35 epoch, train_err: 0.23788622872409648, valid_err: 0.2446996569088503
36 epoch, train_err: 0.2376182269651185, valid_err: 0.24441574235308391
37 epoch, train_err: 0.2373416383388359, valid_err: 0.24412155016269665
38 epoch, train_err: 0.23705584676394104, valid_err: 0.24381652328148606
39 epoch, train_err: 0.23676022749972228, valid_err: 0.24350008440448437
40 epoch, train_err: 0.23645414621371974, valid_err: 0.2431716355531191
41 epoch, train_err: 0.236136958184914, valid_err: 0.24283055774112622
42 epoch, train_err: 0.23580800764602483, valid_err: 0.2424762107408083
43 epoch, train_err: 0.23546662726935894, valid_err: 0.24210793296014163
44 epoch, train_err: 0.23511213780137252, valid_err: 0.24172504144201767
45 epoch, train_err: 0.23474384785168168, valid_err: 0.24132683199753407
46 epoch, train_err: 0.23436105384265396, valid_err: 0.24091257948574152
47 epoch, train_err: 0.23396304012594998, valid_err: 0.24048153825255134
48 epoch, train_err: 0.23354907927242943, valid_err: 0.24003294274161685
49 epoch, train_err: 0.2331184325416882, valid_err: 0.23956600828990535
50 epoch, train_err: 0.23267035053716115, valid_err: 0.23907993212032996
51 epoch, train_err: 0.23220407405217466, valid_err: 0.2385738945432349
52 epoch, train_err: 0.23171883511160019, valid_err: 0.23804706037768356
53 epoch, train_err: 0.23121385821282206, valid_err: 0.237498580602391
54 epoch, train_err: 0.23068836176861604, valid_err: 0.23692759424476723
55 epoch, train_err: 0.23014155975324846, valid_err: 0.23633323051490351
56 epoch, train_err: 0.22957266355167422, valid_err: 0.23571461118944317
57 epoch, train_err: 0.22898088401015992, valid_err: 0.2350708532481687
58 epoch, train_err: 0.2283654336850218, valid_err: 0.23440107176381275
59 epoch, train_err: 0.22772552928448037, valid_err: 0.23370438304312344
60 epoch, train_err: 0.22706039429694044, valid_err: 0.2329799080146021
61 epoch, train_err: 0.22636926179734412, valid_err: 0.2322267758556542
62 epoch, train_err: 0.22565137742166055, valid_err: 0.23144412784918358
63 epoch, train_err: 0.22490600249811174, valid_err: 0.23063112145698803
64 epoch, train_err: 0.22413241732241646, valid_err: 0.2297869345947221
65 epoch, train_err: 0.22332992456320921, valid_err: 0.22891077009071753
66 epoch, train_err: 0.22249785278285117, valid_err: 0.22800186030868352
67 epoch, train_err: 0.2216355600581312, valid_err: 0.2270594719121974
68 epoch, train_err: 0.22074243768482882, valid_err: 0.2260829107470506
69 epoch, train_err: 0.21981791394976008, valid_err: 0.2250715268158654
70 epoch, train_err: 0.21886145795372547, valid_err: 0.2240247193179753
71 epoch, train_err: 0.2178725834686543, valid_err: 0.22294194172632362
72 epoch, train_err: 0.21685085281214161, valid_err: 0.2218227068720288
73 epoch, train_err: 0.21579588072241804, valid_err: 0.22066659200624753
74 epoch, train_err: 0.21470733821649146, valid_err: 0.21947324380795907
75 epoch, train_err: 0.21358495641367362, valid_err: 0.218242383305223
76 epoch, train_err: 0.21242853030586717, valid_err: 0.21697381067626353
77 epoch, train_err: 0.21123792245475992, valid_err: 0.21566740989534072
78 epoch, train_err: 0.21001306659441304, valid_err: 0.21432315318672987
79 epoch, train_err: 0.20875397111558905, valid_err: 0.21294110524822452
80 epoch, train_err: 0.20746072240556018, valid_err: 0.21152142720344153
81 epoch, train_err: 0.2061334880140909, valid_err: 0.21006438023983137
82 epoch, train_err: 0.2047725196128876, valid_err: 0.20857032888683108
83 epoch, train_err: 0.2033781557121651, valid_err: 0.20703974388613872
84 epoch, train_err: 0.2019508240942712, valid_err: 0.20547320460382315
85 epoch, train_err: 0.20049104392071845, valid_err: 0.20387140093213116
86 epoch, train_err: 0.1989994274657593, valid_err: 0.2022351346276377
87 epoch, train_err: 0.1974766814270419, valid_err: 0.20056532003210087
88 epoch, train_err: 0.19592360776220172, valid_err: 0.1988629841232512
89 epoch, train_err: 0.19434110399971513, valid_err: 0.19712926584504842
90 epoch, train_err: 0.1927301629732527, valid_err: 0.19536541467088078
91 epoch, train_err: 0.1910918719313118, valid_err: 0.19357278835892597
92 epoch, train_err: 0.18942741097825766, valid_err: 0.19175284986653496
93 epoch, train_err: 0.18773805080915185, valid_err: 0.1899071634000715
94 epoch, train_err: 0.1860251497089, valid_err: 0.18803738958803135
95 epoch, train_err: 0.18429014979625122, valid_err: 0.18614527977832598
96 epoch, train_err: 0.18253457250483202, valid_err: 0.18423266947505176
97 epoch, train_err: 0.1807600133064582, valid_err: 0.18230147094551863
98 epoch, train_err: 0.1789681356960702, valid_err: 0.18035366504432382
99 epoch, train_err: 0.17716066447237064, valid_err: 0.17839129231733142
100 epoch, train_err: 0.17533937836312563, valid_err: 0.17641644346400667
101 epoch, train_err: 0.17350610205861322, valid_err: 0.17443124925110304
102 epoch, train_err: 0.17166269773034823, valid_err: 0.1724378699836673
103 epoch, train_err: 0.16981105612448297, valid_err: 0.17043848465024472
104 epoch, train_err: 0.16795308732971953, valid_err: 0.16843527986757692
105 epoch, train_err: 0.16609071132778566, valid_err: 0.16643043875570107
106 epoch, train_err: 0.1642258484402158, valid_err: 0.1644261298769263
107 epoch, train_err: 0.16236040978813604, valid_err: 0.1624244963716098
108 epoch, train_err: 0.16049628788189146, valid_err: 0.16042764541997825
109 epoch, train_err: 0.15863534745467184, valid_err: 0.1584376381525951
110 epoch, train_err: 0.15677941664893705, valid_err: 0.15645648012270333
111 epoch, train_err: 0.15493027865661457, valid_err: 0.15448611244191918
112 epoch, train_err: 0.1530896639040591, valid_err: 0.15252840366702666
113 epoch, train_err: 0.1512592428609945, valid_err: 0.15058514251041352
114 epoch, train_err: 0.14944061953952872, valid_err: 0.148658031430475
application/vnd.jupyter.stdout
Output deleted
115 epoch, train_err: 0.14763532573527394, valid_err: 0.14674868114160614
116 epoch, train_err: 0.14584481604809035, valid_err: 0.1448586060667177
117 epoch, train_err: 0.14407046370541648, valid_err: 0.14298922073896658
118 epoch, train_err: 0.1423135571969745, valid_err: 0.14114183714402945
119 epoch, train_err: 0.14057529771619778, valid_err: 0.13931766298009735
120 epoch, train_err: 0.13885679739132747, valid_err: 0.13751780080010706
121 epoch, train_err: 0.13715907827800908, valid_err: 0.1357432479897642
122 epoch, train_err: 0.13548307207556587, valid_err: 0.13399489752576577
123 epoch, train_err: 0.13382962052104644, valid_err: 0.13227353945136935
124 epoch, train_err: 0.1321994764086923, valid_err: 0.1305798630010745
125 epoch, train_err: 0.1305933051776398, valid_err: 0.12891445930260606
126 epoch, train_err: 0.1290116870074104, valid_err: 0.12727782458252834
127 epoch, train_err: 0.12745511935895237, valid_err: 0.1256703638015077
128 epoch, train_err: 0.1259240198985592, valid_err: 0.12409239464632227
129 epoch, train_err: 0.12441872974274294, valid_err: 0.12254415180799755
130 epoch, train_err: 0.12293951696393858, valid_err: 0.12102579147871337
131 epoch, train_err: 0.12148658029956469, valid_err: 0.11953739600420941
132 epoch, train_err: 0.12006005301032066, valid_err: 0.11807897863309247
133 epoch, train_err: 0.1186600068374752, valid_err: 0.11665048830955914
134 epoch, train_err: 0.11728645601315492, valid_err: 0.11525181446141569
135 epoch, train_err: 0.11593936128212215, valid_err: 0.11388279174075397
136 epoch, train_err: 0.11461863389812675, valid_err: 0.11254320468011704
137 epoch, train_err: 0.11332413956249734, valid_err: 0.11123279223232375
138 epoch, train_err: 0.11205570227713667, valid_err: 0.10995125216725853
139 epoch, train_err: 0.11081310808840343, valid_err: 0.10869824530379064
140 epoch, train_err: 0.10959610870246408, valid_err: 0.1074733995595052
141 epoch, train_err: 0.10840442495651763, valid_err: 0.10627631380509403
142 epoch, train_err: 0.10723775013382436, valid_err: 0.105106561514026
143 epoch, train_err: 0.10609575311366871, valid_err: 0.10396369420149673
144 epoch, train_err: 0.10497808135026747, valid_err: 0.10284724464964437
145 epoch, train_err: 0.10388436367718135, valid_err: 0.10175672991862206
146 epoch, train_err: 0.10281421293602126, valid_err: 0.1006916541453523
147 epoch, train_err: 0.10176722843016361, valid_err: 0.09965151113368066
148 epoch, train_err: 0.10074299820582819, valid_err: 0.09863578674121035
149 epoch, train_err: 0.09974110116423529, valid_err: 0.09764396106937462
150 epoch, train_err: 0.09876110900968178, valid_err: 0.09667551046430506
151 epoch, train_err: 0.09780258803926878, valid_err: 0.09572990933682078
152 epoch, train_err: 0.09686510078070767, valid_err: 0.09480663181041157
153 epoch, train_err: 0.09594820748514502, valid_err: 0.09390515320645593
154 epoch, train_err: 0.0950514674823038, valid_err: 0.09302495137611942
155 epoch, train_err: 0.09417444040545939, valid_err: 0.09216550788844162
156 epoch, train_err: 0.09331668729387017, valid_err: 0.0913263090840783
157 epoch, train_err: 0.09247777158029052, valid_err: 0.09050684700401296
158 epoch, train_err: 0.09165725997110954, valid_err: 0.08970662020233423
159 epoch, train_err: 0.09085472322651902, valid_err: 0.08892513445188889
160 epoch, train_err: 0.0900697368479065, valid_err: 0.0881619033512843
161 epoch, train_err: 0.08930188167942997, valid_err: 0.08741644884134978
162 epoch, train_err: 0.08855074443044947, valid_err: 0.08668830163876512
163 epoch, train_err: 0.08781591812519275, valid_err: 0.08597700159416
164 epoch, train_err: 0.08709700248571309, valid_err: 0.08528209798156268
165 epoch, train_err: 0.0863936042538706, valid_err: 0.08460314972565783
166 epoch, train_err: 0.08570533745773731, valid_err: 0.0839397255728935
167 epoch, train_err: 0.0850318236274944, valid_err: 0.0832914042120663
168 epoch, train_err: 0.08437269196556393, valid_err: 0.0826577743496134
169 epoch, train_err: 0.08372757947539633, valid_err: 0.08203843474445155
170 epoch, train_err: 0.08309613105302392, valid_err: 0.08143299420683592
171 epoch, train_err: 0.0824779995451921, valid_err: 0.08084107156535232
172 epoch, train_err: 0.08187284577759157, valid_err: 0.08026229560582308
173 epoch, train_err: 0.08128033855644205, valid_err: 0.07969630498558891
174 epoch, train_err: 0.08070015464641692, valid_err: 0.07914274812632766
175 epoch, train_err: 0.08013197872765492, valid_err: 0.07860128308829345
176 epoch, train_err: 0.07957550333437197, valid_err: 0.07807157742859457
177 epoch, train_err: 0.07903042877737151, valid_err: 0.07755330804588684
178 epoch, train_err: 0.07849646305254727, valid_err: 0.07704616101363217
179 epoch, train_err: 0.07797332173728665, valid_err: 0.07654983140385918
180 epoch, train_err: 0.07746072787650377, valid_err: 0.07606402310317542
181 epoch, train_err: 0.0769584118598724, valid_err: 0.07558844862259388
182 epoch, train_err: 0.07646611129167531, valid_err: 0.07512282890258
183 epoch, train_err: 0.07598357085455004, valid_err: 0.07466689311456945
184 epoch, train_err: 0.075510542168282, valid_err: 0.07422037846007115
185 epoch, train_err: 0.0750467836446783, valid_err: 0.07378302996834535
186 epoch, train_err: 0.07459206033944829, valid_err: 0.07335460029352828
187 epoch, train_err: 0.07414614380191703, valid_err: 0.07293484951197426
188 epoch, train_err: 0.07370881192330878, valid_err: 0.07252354492048839
189 epoch, train_err: 0.0732798487842539, valid_err: 0.07212046083603854
190 epoch, train_err: 0.07285904450209811, valid_err: 0.07172537839745674
191 epoch, train_err: 0.07244619507852443, valid_err: 0.07133808536957004
192 epoch, train_err: 0.0720411022479355, valid_err: 0.07095837595013696
193 epoch, train_err: 0.0716435733269886, valid_err: 0.07058605057991044
194 epoch, train_err: 0.07125342106562382, valid_err: 0.07022091575609368
195 epoch, train_err: 0.07087046349987985, valid_err: 0.069862783849413
196 epoch, train_err: 0.0704945238067503, valid_err: 0.06951147292498787
197 epoch, train_err: 0.07012543016129602, valid_err: 0.06916680656714393
198 epoch, train_err: 0.06976301559619524, valid_err: 0.0688286137082802
199 epoch, train_err: 0.06940711786388185, valid_err: 0.06849672846187439
200 epoch, train_err: 0.06905757930139804, valid_err: 0.06817098995968358
201 epoch, train_err: 0.06871424669805815, valid_err: 0.06785124219317563
202 epoch, train_err: 0.06837697116600472, valid_err: 0.06753733385920582
203 epoch, train_err: 0.0680456080137139, valid_err: 0.06722911820993759
204 epoch, train_err: 0.06772001662249254, valid_err: 0.06692645290698812
205 epoch, train_err: 0.0674000603259935, valid_err: 0.0666291998797705
206 epoch, train_err: 0.06708560629276208, valid_err: 0.06633722518798807
207 epoch, train_err: 0.06677652541181506, valid_err: 0.0660503988882302
208 epoch, train_err: 0.06647269218124338, valid_err: 0.06576859490460962
209 epoch, train_err: 0.06617398459982049, valid_err: 0.0654916909033737
210 epoch, train_err: 0.06588028406158997, valid_err: 0.06521956817141521
211 epoch, train_err: 0.0655914752534002, valid_err: 0.06495211149860775
212 epoch, train_err: 0.06530744605534763, valid_err: 0.06468920906388051
213 epoch, train_err: 0.0650280874440845, valid_err: 0.06443075232494927
214 epoch, train_err: 0.06475329339894416, valid_err: 0.06417663591161625
215 epoch, train_err: 0.06448296081083219, valid_err: 0.06392675752254849
216 epoch, train_err: 0.06421698939382926, valid_err: 0.06368101782544561
217 epoch, train_err: 0.06395528159944942, valid_err: 0.0634393203605048
218 epoch, train_err: 0.06369774253349572, valid_err: 0.0632015714470932
219 epoch, train_err: 0.06344427987545274, valid_err: 0.0629676800935344
220 epoch, train_err: 0.06319480380035636, valid_err: 0.0627375579099208
221 epoch, train_err: 0.062949226903078, valid_err: 0.06251111902385956
222 epoch, train_err: 0.06270746412496278, valid_err: 0.06228827999906469
223 epoch, train_err: 0.06246943268275875, valid_err: 0.062068959756706475
224 epoch, train_err: 0.062235051999775096, valid_err: 0.06185307949943309
225 epoch, train_err: 0.062004243639209894, valid_err: 0.06164056263797721
226 epoch, train_err: 0.06177693123958273, valid_err: 0.06143133472026736
227 epoch, train_err: 0.06155304045221526, valid_err: 0.06122532336295936
228 epoch, train_err: 0.061332498880697875, valid_err: 0.061022458185310706
application/vnd.jupyter.stdout
Output deleted
229 epoch, train_err: 0.061115236022284615, valid_err: 0.060822670745318434
230 epoch, train_err: 0.06090118321115798, valid_err: 0.06062589447804511
231 epoch, train_err: 0.06069027356350764, valid_err: 0.06043206463605946
232 epoch, train_err: 0.060482441924366484, valid_err: 0.06024111823191862
233 epoch, train_err: 0.06027762481614996, valid_err: 0.06005299398262328
234 epoch, train_err: 0.06007576038884549, valid_err: 0.059867632255977236
235 epoch, train_err: 0.059876788371800084, valid_err: 0.059684975018784794
236 epoch, train_err: 0.05968065002705452, valid_err: 0.05950496578682334
237 epoch, train_err: 0.059487288104175945, valid_err: 0.059327549576528645
238 epoch, train_err: 0.0592966467965393, valid_err: 0.059152672858331555
239 epoch, train_err: 0.059108671699011434, valid_err: 0.0589802835115904
240 epoch, train_err: 0.05892330976699236, valid_err: 0.05881033078106101
241 epoch, train_err: 0.05874050927676918, valid_err: 0.058642765234850014
242 epoch, train_err: 0.058560219787138906, valid_err: 0.05847753872379999
243 epoch, train_err: 0.05838239210225992, valid_err: 0.05831460434225457
244 epoch, train_err: 0.058206978235689655, valid_err: 0.05815391639015477
245 epoch, train_err: 0.05803393137557106, valid_err: 0.057995430336419386
246 epoch, train_err: 0.057863205850928065, valid_err: 0.05783910278356272
247 epoch, train_err: 0.05769475709903432, valid_err: 0.05768489143350691
248 epoch, train_err: 0.05752854163381874, valid_err: 0.05753275505454499
249 epoch, train_err: 0.05736451701527365, valid_err: 0.057382653449414144
250 epoch, train_err: 0.05720264181983167, valid_err: 0.05723454742443912
251 epoch, train_err: 0.05704287561167921, valid_err: 0.05708839875970826
252 epoch, train_err: 0.056885178914975125, valid_err: 0.056944170180243836
253 epoch, train_err: 0.056729513186943914, valid_err: 0.05680182532813336
254 epoch, train_err: 0.05657584079181473, valid_err: 0.056661328735584986
255 epoch, train_err: 0.056424124975577035, valid_err: 0.05652264579887642
256 epoch, train_err: 0.05627432984152648, valid_err: 0.05638574275316353
257 epoch, train_err: 0.05612642032657345, valid_err: 0.05625058664811923
258 epoch, train_err: 0.05598036217828956, valid_err: 0.056117145324372814
259 epoch, train_err: 0.05583612193266699, valid_err: 0.05598538739072099
260 epoch, train_err: 0.05569366689256699, valid_err: 0.0558552822020836
261 epoch, train_err: 0.05555296510683405, valid_err: 0.05572679983817723
262 epoch, train_err: 0.05541398535005377, valid_err: 0.05559991108288145
263 epoch, train_err: 0.055276697102933585, valid_err: 0.05547458740427378
264 epoch, train_err: 0.05514107053328384, valid_err: 0.055350800935308114
265 epoch, train_err: 0.055007076477581344, valid_err: 0.055228524455116335
266 epoch, train_err: 0.05487468642309419, valid_err: 0.055107731370909005
267 epoch, train_err: 0.05474387249055045, valid_err: 0.0549883957004553
268 epoch, train_err: 0.05461460741733174, valid_err: 0.05487049205512142
269 epoch, train_err: 0.05448686454117491, valid_err: 0.0547539956234484
270 epoch, train_err: 0.054360617784365045, valid_err: 0.05463888215524984
271 epoch, train_err: 0.054235841638403205, valid_err: 0.05452512794621217
272 epoch, train_err: 0.054112511149133145, valid_err: 0.05441270982297935
273 epoch, train_err: 0.05399060190231243, valid_err: 0.054301605128705265
274 epoch, train_err: 0.05387009000961306, valid_err: 0.05419179170905858
275 epoch, train_err: 0.053750952095037864, valid_err: 0.054083247898663006
276 epoch, train_err: 0.05363316528173777, valid_err: 0.05397595250795889
277 epoch, train_err: 0.053516707179219046, valid_err: 0.05386988481047157
278 epoch, train_err: 0.0534015558709259, valid_err: 0.05376502453047239
279 epoch, train_err: 0.05328768990218705, valid_err: 0.053661351831019076
280 epoch, train_err: 0.053175088268514505, valid_err: 0.053558847302362045
281 epoch, train_err: 0.053063730404243045, valid_err: 0.05345749195070557
282 epoch, train_err: 0.05295359617149926, valid_err: 0.05335726718730965
283 epoch, train_err: 0.05284466584948959, valid_err: 0.05325815481792326
284 epoch, train_err: 0.05273692012409758, valid_err: 0.05316013703253625
285 epoch, train_err: 0.0526303400777796, valid_err: 0.05306319639543996
286 epoch, train_err: 0.05252490717975033, valid_err: 0.052967315835585824
287 epoch, train_err: 0.052420603276448124, valid_err: 0.052872478637232136
288 epoch, train_err: 0.05231741058227172, valid_err: 0.05277866843086944
289 epoch, train_err: 0.05221531167057982, valid_err: 0.05268586918441559
290 epoch, train_err: 0.05211428946494475, valid_err: 0.05259406519467008
291 epoch, train_err: 0.052014327230652385, valid_err: 0.052503241079021545
292 epoch, train_err: 0.05191540856644099, valid_err: 0.05241338176739739
293 epoch, train_err: 0.05181751739647104, valid_err: 0.052324472494450075
294 epoch, train_err: 0.05172063796251852, valid_err: 0.05223649879197002
295 epoch, train_err: 0.05162475481638559, valid_err: 0.05214944648151957
296 epoch, train_err: 0.05152985281252158, valid_err: 0.05206330166728021
297 epoch, train_err: 0.05143591710084667, valid_err: 0.05197805072910484
298 epoch, train_err: 0.051342933119773906, valid_err: 0.05189368031577141
299 epoch, train_err: 0.051250886589421875, valid_err: 0.051810177338427445
300 epoch, train_err: 0.05115976350501306, valid_err: 0.05172752896422371
301 epoch, train_err: 0.05106955013045129, valid_err: 0.05164572261012683
302 epoch, train_err: 0.05098023299207444, valid_err: 0.05156474593690837
303 epoch, train_err: 0.05089179887257491, valid_err: 0.051484586843302614
304 epoch, train_err: 0.05080423480508425, valid_err: 0.051405233460328874
305 epoch, train_err: 0.05071752806741711, valid_err: 0.05132667414577259
306 epoch, train_err: 0.05063166617646802, valid_err: 0.0512488974788203
307 epoch, train_err: 0.05054663688275863, valid_err: 0.05117189225484397
308 epoch, train_err: 0.05046242816512923, valid_err: 0.05109564748032922
309 epoch, train_err: 0.05037902822557074, valid_err: 0.05102015236794382
310 epoch, train_err: 0.05029642548419342, valid_err: 0.05094539633174175
311 epoch, train_err: 0.05021460857432759, valid_err: 0.05087136898249801
312 epoch, train_err: 0.05013356633775261, valid_err: 0.050798060123171486
313 epoch, train_err: 0.05005328782005045, valid_err: 0.05072545974449016
314 epoch, train_err: 0.049973762266080725, valid_err: 0.050653558020657075
315 epoch, train_err: 0.04989497911557194, valid_err: 0.0505823453051707
316 epoch, train_err: 0.04981692799882734, valid_err: 0.05051181212675857
317 epoch, train_err: 0.04973959873254152, valid_err: 0.050441949185419915
318 epoch, train_err: 0.049662981315723734, valid_err: 0.050372747348573006
319 epoch, train_err: 0.049587065925725884, valid_err: 0.05030419764730547
320 epoch, train_err: 0.04951184291437139, valid_err: 0.05023629127272404
321 epoch, train_err: 0.049437302804182764, valid_err: 0.05016901957240021
322 epoch, train_err: 0.04936343628470421, valid_err: 0.050102374046909225
323 epoch, train_err: 0.04929023420891722, valid_err: 0.050036346346460286
324 epoch, train_err: 0.049217687589745866, valid_err: 0.049970928267613894
325 epoch, train_err: 0.049145787596650156, valid_err: 0.04990611175008523
326 epoch, train_err: 0.049074525552303995, valid_err: 0.04984188887362971
327 epoch, train_err: 0.049003892929356, valid_err: 0.04977825185500939
328 epoch, train_err: 0.048933881347269996, valid_err: 0.049715193045036826
329 epoch, train_err: 0.04886448256924457, valid_err: 0.04965270492569516
330 epoch, train_err: 0.04879568849920723, valid_err: 0.04959078010733127
331 epoch, train_err: 0.048727491178883406, valid_err: 0.049529411325920376
332 epoch, train_err: 0.048659882784936284, valid_err: 0.049468591440400146
333 epoch, train_err: 0.04859285562617656, valid_err: 0.04940831343007176
334 epoch, train_err: 0.04852640214084022, valid_err: 0.04934857039206641
335 epoch, train_err: 0.048460514893931585, valid_err: 0.04928935553887521
336 epoch, train_err: 0.04839518657463075, valid_err: 0.049230662195941015
337 epoch, train_err: 0.04833040999376358, valid_err: 0.049172483799309566
338 epoch, train_err: 0.048266178081331504, valid_err: 0.04911481389333943
339 epoch, train_err: 0.04820248388410104, valid_err: 0.04905764612846791
340 epoch, train_err: 0.048139320563250154, valid_err: 0.049000974259032354
341 epoch, train_err: 0.048076681392070615, valid_err: 0.048944792141144326
342 epoch, train_err: 0.048014559753724435, valid_err: 0.04888909373061619
application/vnd.jupyter.stdout
Output deleted
343 epoch, train_err: 0.04795294913905333, valid_err: 0.04883387308093781
344 epoch, train_err: 0.047891843144439304, valid_err: 0.048779124341302244
345 epoch, train_err: 0.047831235469715544, valid_err: 0.048724841754679185
346 epoch, train_err: 0.047771119916125725, valid_err: 0.04867101965593483
347 epoch, train_err: 0.0477114903843307, valid_err: 0.048617652469996525
348 epoch, train_err: 0.047652340872461646, valid_err: 0.0485647347100615
349 epoch, train_err: 0.04759366547421765, valid_err: 0.04851226097584829
350 epoch, train_err: 0.04753545837700745, valid_err: 0.048460225951889105
351 epoch, train_err: 0.04747771386013323, valid_err: 0.04840862440586293
352 epoch, train_err: 0.04742042629301645, valid_err: 0.04835745118696797
353 epoch, train_err: 0.04736359013346378, valid_err: 0.04830670122433171
354 epoch, train_err: 0.047307199925971974, valid_err: 0.048256369525458746
355 epoch, train_err: 0.047251250300071906, valid_err: 0.048206451174714296
356 epoch, train_err: 0.04719573596870917, valid_err: 0.04815694133184288
357 epoch, train_err: 0.0471406517266613, valid_err: 0.04810783523052163
358 epoch, train_err: 0.04708599244899053, valid_err: 0.048059128176946464
359 epoch, train_err: 0.04703175308953075, valid_err: 0.048010815548450915
360 epoch, train_err: 0.046977928679408476, valid_err: 0.047962892792156424
361 epoch, train_err: 0.04692451432559619, valid_err: 0.04791535542365349
362 epoch, train_err: 0.046871505209498024, valid_err: 0.04786819902571254
363 epoch, train_err: 0.04681889658556607, valid_err: 0.04782141924702435
364 epoch, train_err: 0.046766683779947554, valid_err: 0.04777501180096808
365 epoch, train_err: 0.04671486218916151, valid_err: 0.04772897246440808
366 epoch, train_err: 0.0466634272788037, valid_err: 0.047683297076516536
367 epoch, train_err: 0.04661237458228056, valid_err: 0.04763798153762289
368 epoch, train_err: 0.04656169969956983, valid_err: 0.04759302180808876
369 epoch, train_err: 0.046511398296008234, valid_err: 0.04754841390720772
370 epoch, train_err: 0.0464614661011054, valid_err: 0.047504153912129296
371 epoch, train_err: 0.04641189890738274, valid_err: 0.047460237956806824
372 epoch, train_err: 0.04636269256923766, valid_err: 0.0474166622309678
373 epoch, train_err: 0.04631384300183148, valid_err: 0.0473734229791072
374 epoch, train_err: 0.04626534618000136, valid_err: 0.047330516499502845
375 epoch, train_err: 0.046217198137195034, valid_err: 0.04728793914325118
376 epoch, train_err: 0.046169394964428054, valid_err: 0.04724568731332507
377 epoch, train_err: 0.046121932809263015, valid_err: 0.04720375746365076
378 epoch, train_err: 0.04607480787481021, valid_err: 0.0471621460982054
379 epoch, train_err: 0.046028016418748825, valid_err: 0.04712084977013348
380 epoch, train_err: 0.04598155475236897, valid_err: 0.04707986508088205
381 epoch, train_err: 0.04593541923963347, valid_err: 0.04703918867935452
382 epoch, train_err: 0.045889606296258864, valid_err: 0.04699881726108201
383 epoch, train_err: 0.045844112388815444, valid_err: 0.046958747567412536
384 epoch, train_err: 0.045798934033846106, valid_err: 0.04691897638471679
385 epoch, train_err: 0.045754067797002716, valid_err: 0.046879500543610966
386 epoch, train_err: 0.04570951029220068, valid_err: 0.04684031691819526
387 epoch, train_err: 0.045665258180790196, valid_err: 0.04680142242530874
388 epoch, train_err: 0.04562130817074487, valid_err: 0.04676281402379931
389 epoch, train_err: 0.04557765701586632, valid_err: 0.04672448871380869
390 epoch, train_err: 0.045534301515004955, valid_err: 0.046686443536072236
391 epoch, train_err: 0.045491238511296585, valid_err: 0.04664867557123307
392 epoch, train_err: 0.04544846489141424, valid_err: 0.04661118193917026
393 epoch, train_err: 0.04540597758483479, valid_err: 0.04657395979834052
394 epoch, train_err: 0.04536377356312036, valid_err: 0.046537006345133136
395 epoch, train_err: 0.04532184983921407, valid_err: 0.04650031881323853
396 epoch, train_err: 0.04528020346674937, valid_err: 0.04646389447302876
397 epoch, train_err: 0.04523883153937358, valid_err: 0.04642773063095098
398 epoch, train_err: 0.04519773119008416, valid_err: 0.04639182462893267
399 epoch, train_err: 0.045156899590578455, valid_err: 0.046356173843799135
400 epoch, train_err: 0.0451163339506158, valid_err: 0.046320775686702155
401 epoch, train_err: 0.045076031517392326, valid_err: 0.046285627602560074
402 epoch, train_err: 0.04503598957492791, valid_err: 0.04625072706950923
403 epoch, train_err: 0.04499620544346473, valid_err: 0.04621607159836561
404 epoch, train_err: 0.04495667647887783, valid_err: 0.046181658732097466
405 epoch, train_err: 0.04491740007209666, valid_err: 0.04614748604530807
406 epoch, train_err: 0.04487837364853816, valid_err: 0.04611355114372864
407 epoch, train_err: 0.04483959466755018, valid_err: 0.04607985166372105
408 epoch, train_err: 0.044801060621865965, valid_err: 0.046046385271789926
409 epoch, train_err: 0.04476276903706891, valid_err: 0.04601314966410459
410 epoch, train_err: 0.044724717471067287, valid_err: 0.04598014256602966
411 epoch, train_err: 0.044686903513579096, valid_err: 0.04594736173166522
412 epoch, train_err: 0.0446493247856266, valid_err: 0.04591480494339519
413 epoch, train_err: 0.04461197893904031, valid_err: 0.04588247001144467
414 epoch, train_err: 0.044574863655972345, valid_err: 0.045850354773445214
415 epoch, train_err: 0.04453797664841885, valid_err: 0.04581845709400912
416 epoch, train_err: 0.044501315657751256, valid_err: 0.04578677486431104
417 epoch, train_err: 0.044464878454256576, valid_err: 0.04575530600167722
418 epoch, train_err: 0.04442866283668553, valid_err: 0.04572404844918345
419 epoch, train_err: 0.04439266663180985, valid_err: 0.04569300017525944
420 epoch, train_err: 0.044356887693987164, valid_err: 0.04566215917330134
421 epoch, train_err: 0.044321323904734006, valid_err: 0.045631523461290735
422 epoch, train_err: 0.04428597317230683, valid_err: 0.045601091081421165
423 epoch, train_err: 0.04425083343129048, valid_err: 0.045570860099731304
424 epoch, train_err: 0.04421590264219415, valid_err: 0.04554082860574505
425 epoch, train_err: 0.044181178791054924, valid_err: 0.045510994712117926
426 epoch, train_err: 0.0441466598890481, valid_err: 0.04548135655429004
427 epoch, train_err: 0.04411234397210489, valid_err: 0.04545191229014549
428 epoch, train_err: 0.044078229100536874, valid_err: 0.04542266009967757
429 epoch, train_err: 0.044044313358667146, valid_err: 0.04539359818466043
430 epoch, train_err: 0.044010594854468135, valid_err: 0.045364724768326405
431 epoch, train_err: 0.043977071719205965, valid_err: 0.04533603809504909
432 epoch, train_err: 0.04394374210709085, valid_err: 0.04530753643003232
433 epoch, train_err: 0.04391060419493411, valid_err: 0.045279218059004214
434 epoch, train_err: 0.04387765618181084, valid_err: 0.045251081287917216
435 epoch, train_err: 0.04384489628872886, valid_err: 0.045223124442653216
436 epoch, train_err: 0.043812322758303386, valid_err: 0.04519534586873377
437 epoch, train_err: 0.043779933854437285, valid_err: 0.045167743931035134
438 epoch, train_err: 0.043747727862007126, valid_err: 0.04514031701350926
439 epoch, train_err: 0.0437157030865545, valid_err: 0.045113063518908536
440 epoch, train_err: 0.04368385785398293, valid_err: 0.04508598186851622
441 epoch, train_err: 0.043652190510259786, valid_err: 0.04505907050188076
442 epoch, train_err: 0.04362069942112372, valid_err: 0.045032327876555416
443 epoch, train_err: 0.04358938297179685, valid_err: 0.045005752467841914
444 epoch, train_err: 0.04355823956670186, valid_err: 0.044979342768538794
445 epoch, train_err: 0.04352726762918421, valid_err: 0.04495309728869376
446 epoch, train_err: 0.04349646560123889, valid_err: 0.044927014555360655
447 epoch, train_err: 0.043465831943242085, valid_err: 0.044901093112360325
448 epoch, train_err: 0.04343536513368701, valid_err: 0.044875331520045254
449 epoch, train_err: 0.04340506366892467, valid_err: 0.04484972835506918
450 epoch, train_err: 0.04337492606290862, valid_err: 0.04482428221015908
451 epoch, train_err: 0.04334495084694429, valid_err: 0.04479899169389272
452 epoch, train_err: 0.04331513656944259, valid_err: 0.04477385543047859
453 epoch, train_err: 0.04328548179567723, valid_err: 0.04474887205953977
454 epoch, train_err: 0.043255985107546656, valid_err: 0.04472404023590233
455 epoch, train_err: 0.0432266451033396, valid_err: 0.04469935862938588
456 epoch, train_err: 0.043197460397504585, valid_err: 0.04467482592459865
457 epoch, train_err: 0.04316842962042343, valid_err: 0.04465044082073538
458 epoch, train_err: 0.04313955141818825, valid_err: 0.044626202031378764
application/vnd.jupyter.stdout
Output deleted
459 epoch, train_err: 0.04311082445238247, valid_err: 0.044602108284304186
460 epoch, train_err: 0.04308224739986503, valid_err: 0.04457815832128749
461 epoch, train_err: 0.04305381895255854, valid_err: 0.04455435089791604
462 epoch, train_err: 0.04302553781724068, valid_err: 0.044530684783402726
463 epoch, train_err: 0.04299740271533912, valid_err: 0.04450715876040315
464 epoch, train_err: 0.04296941238272956, valid_err: 0.04448377162483561
465 epoch, train_err: 0.042941565569537495, valid_err: 0.04446052218570417
466 epoch, train_err: 0.04291386103994274, valid_err: 0.0444374092649244
467 epoch, train_err: 0.04288629757198736, valid_err: 0.04441443169715179
468 epoch, train_err: 0.04285887395738658, valid_err: 0.04439158832961351
469 epoch, train_err: 0.042831589001343005, valid_err: 0.044368878021942004
470 epoch, train_err: 0.04280444152236346, valid_err: 0.044346299646011944
471 epoch, train_err: 0.042777430352078906, valid_err: 0.04432385208577952
472 epoch, train_err: 0.04275055433506746, valid_err: 0.04430153423712395
473 epoch, train_err: 0.04272381232867987, valid_err: 0.044279345007692025
474 epoch, train_err: 0.04269720320286791, valid_err: 0.04425728331674489
475 epoch, train_err: 0.04267072584001568, valid_err: 0.044235348095006996
476 epoch, train_err: 0.04264437913477322, valid_err: 0.04421353828451763
477 epoch, train_err: 0.04261816199389306, valid_err: 0.04419185283848497
478 epoch, train_err: 0.042592073336069236, valid_err: 0.04417029072114183
479 epoch, train_err: 0.04256611209177874, valid_err: 0.04414885090760425
480 epoch, train_err: 0.04254027720312551, valid_err: 0.04412753238373191
481 epoch, train_err: 0.042514567623687036, valid_err: 0.04410633414599075
482 epoch, train_err: 0.042488982318363046, valid_err: 0.044085255201317716
483 epoch, train_err: 0.04246352026322674, valid_err: 0.04406429456698776
484 epoch, train_err: 0.042438180445378365, valid_err: 0.044043451270482564
485 epoch, train_err: 0.04241296186280084, valid_err: 0.044022724349361386
486 epoch, train_err: 0.04238786352421781, valid_err: 0.044002112851133986
487 epoch, train_err: 0.042362884448953815, valid_err: 0.043981615833135486
488 epoch, train_err: 0.04233802366679667, valid_err: 0.04396123236240263
489 epoch, train_err: 0.04231328021786182, valid_err: 0.04394096151555286
490 epoch, train_err: 0.042288653152458736, valid_err: 0.04392080237866405
491 epoch, train_err: 0.042264141530959656, valid_err: 0.04390075404715723
492 epoch, train_err: 0.04223974442367015, valid_err: 0.04388081562567989
493 epoch, train_err: 0.042215460910701295, valid_err: 0.04386098622799202
494 epoch, train_err: 0.0421912900818444, valid_err: 0.04384126497685343
495 epoch, train_err: 0.042167231036447116, valid_err: 0.04382165100391228
496 epoch, train_err: 0.042143282883291944, valid_err: 0.04380214344959634
497 epoch, train_err: 0.04211944474047578, valid_err: 0.04378274146300475
498 epoch, train_err: 0.042095715735292145, valid_err: 0.043763444201802094
499 epoch, train_err: 0.04207209500411444, valid_err: 0.04374425083211373
TESZT
0.02668529887033235
elapsed time:0:00:18.205375
application/vnd.jupyter.stdout
In [17]:
xxxxxxxxxx
1
 
1
%tensorboard --logdir logs
Metadata changed
xxxxxxxxxx
In [17]:
xxxxxxxxxx
5
 
1
##these are for jupyter notebook applications#
2
#!taskkill /im tensorboard.exe /f
3
#!del /q %TMP%\.tensorboard-info\*
4
#############################################
5
%tensorboard --logdir logs --host localhost
Metadata changed
xxxxxxxxxx
Reusing TensorBoard on port 6006 (pid 2440), started 0:00:53 ago. (Use '!kill 2440' to kill it.)
text/plain